In any case i think a default loading state should not include any html or css. I suspect there are too many ui edge cases. layout shifts, theming, writing direction, etc.
02.09.2025 23:38 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0@tryggvigy.bsky.social
Dev @ spotify, desktop client
In any case i think a default loading state should not include any html or css. I suspect there are too many ui edge cases. layout shifts, theming, writing direction, etc.
02.09.2025 23:38 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0What i meant by that is if we are transitioning from nothing to something in ui, then a good default would be to have no loading state at all. Maybe transitions in react only handle updates to existing to ui, im not sure! Isn't this more or less how things are today?
02.09.2025 23:27 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0Feels like a good default would be null if nothing was there before, and the previous value if something was there before
02.09.2025 23:00 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0it could taste like chicken. Only one way to find out www.wildfooduk.com/mushroom-gui...
27.08.2025 14:53 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0proof server_responded ๐ฅฒ I'm curious how the messiness of the real world would be modelled. Like Haskel? In any case the type system in Lean seems real powerful and useful
19.08.2025 00:53 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0@bram.us do you know if Chromium has addressed this already?
14.08.2025 15:40 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0Oh really! I'm not gonna pretend i understand half of this stuff ๐ We were hit with this bad just a couple of months ago. I'll have to test this again as well. There be dragons here at least so be careful! Var substitution on the compositor is probably not implemented in all baseline browsers.
14.08.2025 15:35 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0This is real useful, thanks for sharing! Folks seeing the var() in keyframes just be aware of this www.bram.us/2023/02/01/t... they can be a real performance gotcha in larger apps
14.08.2025 00:33 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0Wrote about some of the gotchas with CSS @layer blog-tryggvi.fly.dev/blog/css-lay...
07.08.2025 11:00 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0Marquees.. Love them or hate them. Probably the latter! There are niche places for them, but are tricky to implement in an accessible and performant way on the web. blog-tryggvi.fly.dev/blog/perform...
06.08.2025 16:24 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0Wrote my first blog post yesterday! blog-tryggvi.fly.dev/blog/thought...
04.08.2025 16:35 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0Bluesky doesn't penalize you for using links, tags, using key words like Patreon or commissions, and doesn't have a central algorithm on your following feed, which means all the power rests on you to help your favourite creators be seen!
30.07.2025 18:05 โ ๐ 15877 ๐ 6409 ๐ฌ 11 ๐ 1Finally some improvements to GitHub review ui! I've been unhappy with it for so many years! I really hope it's virtual scrolled now, that's just table stakes in 2025.
github.blog/changelog/20...
Happy "European Accessibility Act goes into effect" week to all who celebrate
25.06.2025 17:36 โ ๐ 140 ๐ 37 ๐ฌ 0 ๐ 1Yeah that's true. Especially with the recent interop changes where the line is becoming blurred
16.06.2025 06:28 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0started this project esmchecker.com a while ago to help authors check how ESM ready projects are. Also available as a CLI tool to check repos not on npm.
15.06.2025 22:20 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0Great article! I havent tried RCS yet but the article brought up questions for me like: "can I dynamically import a 'use client' module on the server?" And "how spec compatible with esm will this be? Especially going forward with new additions like Module Blocks www.proposals.es/proposals/Mo... etc"
06.06.2025 18:33 โ ๐ 1 ๐ 0 ๐ฌ 0 ๐ 0It is magnificent! I expanded it to work in the block direction as well in codepen.io/tryggvigy/pe... with line-clamp support too. CSS is on fire
29.05.2025 20:37 โ ๐ 2 ๐ 0 ๐ฌ 0 ๐ 0Awesome! Also, that has to be the most outrageous pun I've ever seen that you did there, hats off to you
28.05.2025 06:51 โ ๐ 5 ๐ 0 ๐ฌ 1 ๐ 0esmchecker.com also works now ๐
11.05.2025 20:45 โ ๐ 1 ๐ 0 ๐ฌ 0 ๐ 0Try it out, it's super fast ๐ฆ! And let me know if you find any resolve/parse errors
11.05.2025 20:07 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0New release of github.com/tryggvigy/es... is out! Now also available via npx ๐
Try it on your repo:
`npx @esm-checker/checker -p ./package.json`
Or analyze any npm package at:
esm-checker.fly.dev
and it's FAST! I ran it on a repo with 357 dependencies and it takes less than 1 second
11.05.2025 11:40 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0`@esm-checker/checker` is a CLI/JS lib version of esm-checker.fly.dev that can be run on any repo to create a report for all the dependencies in package.json
11.05.2025 11:34 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0Finally got the cross compilation and npm publishing of @esm-checker/checker working in Github Actions! It's getting close to being able to be used by anyone to check a whole repo. Created a few issues to follow up github.com/tryggvigy/es... also need to write docs
11.05.2025 11:33 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0yeah that could be good to support, if "exports" are specified with multiple entry-points to allow just checking one. Don't think that should be hard to add
08.05.2025 07:59 โ ๐ 1 ๐ 0 ๐ฌ 0 ๐ 0I think a good default is to not install optional peers, but having an option to turn on installing them would be a good feature. I can imagine there are many packages out there with subpath exports that are only meant to be used with an optional peer dep
08.05.2025 07:16 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0I'll have to think about that! Would be good to have deterministic order of failures. It's a bit hard since I'm parallelizing everything heavily
08.05.2025 07:14 โ ๐ 1 ๐ 0 ๐ฌ 0 ๐ 0At first glance it actually makes sense to me that this could be happening here! This esm dist file is importing "react" but convex specifies react as an optional peer dependency so this resolution could fail? Maybe im missing something
08.05.2025 05:44 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0It's way more likely that I have a bug! It's still very rough around the edges
08.05.2025 04:53 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0